Distance Dependent Chinese Restaurant Processes

Abstract

We develop the distance dependent Chinese restaurant process (CRP), a flexible class of distributions over partitions that allows for non-exchangeability. This class can be used to model many kinds of dependencies between data in infinite clustering models, including dependencies across time or space. We examine the properties of the distance dependent CRP, discuss its connections to Bayesian nonparametric mixture models, and derive a Gibbs sampler for both observed and mixture settings. We study its performance with three text corpora. We show that relaxing the assumption of exchangeability with distance dependent CRPs can provide a better fit to sequential data. We also show its alternative formulation of the traditional CRP leads to a faster-mixing Gibbs sampling algorithm than the one based on the original formulation.

References in corpus (2)

• Nonparametric Bayes Pachinko Allocation
• Fast search for Dirichlet process mixture models

Cited by in corpus (6)

• A Split-Merge MCMC Algorithm for the Hierarchical Dirichlet Process
• Bayesian Inference with Posterior Regularization and applications to Infinite Latent SVMs
• Big Learning with Bayesian Methods
• Distance Dependent Infinite Latent Feature Models
• Dynamic Multi-Relational Chinese Restaurant Process for Analyzing Influences on Users in Social Media
• Comparative Judgement Modeling to Map Forced Marriage at Local Levels